Upload & Sell: On
| Re: Why Do Canon Lenses Score so Low? |
Most published tests of lenses usually leave out too much, especially differences in how different lenses render color and contrast. Sharpness is a subjective impression, not equivalent to resolution. Resolution is objectively measurable at least in principle. But resolution measured with a black and white test target can't show the ability of a lens to distinguish subtle shadings among close-together colors. Nor does it show very well whether the lens suffers color blindness in parts of the spectrum, nor what color casts it delivers. Good performance in those areas adds much to the subjective impression of sharpness, but tends to go unmeasured in published tests.
For that reason, published test results tend to be more useful for weeding out clunkers than for distinguishing among better-serviceable lenses. Among the latter, adequate resolution is the rule, but unmeasured differences in color and contrast nevertheless manifest as notable differences in the subjective impression of sharpness.
To take an example, the Canon 50mm f/1.4 delivers high resolution when used at f/5.6. The excellent and generally conservative Photozone test site rates it very well. At that aperture, the site gives the Canon a slight resolution edge over the Zeiss 50 f/2.0 MP. But they are close, and Photozone's charts compare resolution without measuring color rendering (except for aberrations, measured separately from resolution).
In side-by-side image comparisons that I make myself, I rate the Zeiss as subjectively sharper, and the difference does not seem particularly close. Better Zeiss color and contrast account for the difference. I doubt many image viewers who enjoy normal color vision would reach any different conclusion. Advantageously rendered subtle colors make the Zeiss images at least seem to show more detail. I suspect the difference is not merely subjective, but real, and would be readily measurable by methods different than those used to measure resolution.